DE-NOISING AND OPTIMIZATION OF MEDICAL IMAGES USING DEEP LEARNING TECHNIQUES
Advanced image securing frameworks envelop various optical and electronic gadgets whose deficiencies cause clamor in the gained image. Clinical image handling involves algorithms and systems for tasks like image upgradation, compression of images, and so on. A few cutting-edge systems that convert human organs into computerized images for treatment incorporates X ray, Computed Tomography (CT), Magnetic Resonance Imaging (MRI). The major drawback of these medical images acquired by the said modalities is the errors in the images because of image acquisition and alteration in the images. These errors in images are called as noise. So, all the clinical images require some kind of De-noising techniques to improve the quality of the images being considered by experts to perform analysis and diagnosis. This work proposes an efficient De-noising approach by using Recurrent Neural network with Long-short-term memory (LSTM) and genetic optimization to eliminate Gaussian, and Salt & Pepper Noises.
De-Noising of Medical Images; Image Noises; Optimization of Images; Deep Learning Techniques; Batch Normalization; Recurrent Neural Networks; Long Short Term Memory (LSTM)